PaperlayerView exact revision on arXiv

Source-linked AI summary

Self-Control in Cyberspace: Applying Dual Systems Theory to a Review of Digital Self-Control Tools

Ulrik Lyngs, Kai Lukoff, Petr Slovak, Reuben Binns, Adam Slack, Michael Inzlicht, Max Van Kleek, Nigel Shadbolt

arXiv:1902.00157v1cs.HC

TL;DR

People commonly struggle to control digital-device use, while the design mechanisms supporting self-control remain insufficiently understood. The paper reviews 367 apps and browser extensions and applies an integrative dual systems model with expected value of control to organise their features. It uses this framework to clarify how existing tools support self-control and to identify underexplored cognitive mechanisms for future designs.

  • Problem

    Research has not systematically mapped design features across existing digital self-control tools, despite widespread struggles with device use and limited understanding of effective approaches.

  • Method

    The paper reviews 367 apps and browser extensions from three stores and applies an integrative dual systems model incorporating expected value of control.

  • Results

    The analysis organises common design features and intervention strategies according to dual systems theory, including mechanisms related to fluctuating self-control.

  • Takeaways & Limitations

    The model clarifies how tools may scaffold self-control and reveals underexplored cognitive mechanisms that could inform new tool designs.

  • Takeaways & Limitations

    System 2 control is limited by capacity and fluctuates with fatigue and emotional state.

Abstract

from arXiv · show

Many people struggle to control their use of digital devices. However, our understanding of the design mechanisms that support user self-control remains limited. In this paper, we make two contributions to HCI research in this space: first, we analyse 367 apps and browser extensions from the Google Play, Chrome Web, and Apple App stores to identify common core design features and intervention strategies afforded by current tools for digital self-control. Second, we adapt and apply an integrative dual systems model of self-regulation as a framework for organising and evaluating the design features found. Our analysis aims to help the design of better tools in two ways: (i) by identifying how, through a well-established model of self-regulation, current tools overlap and differ in how they support self-control; and (ii) by using the model to reveal underexplored cognitive mechanisms that could aid the design of new tools.

INTRODUCTION

Digital devices offer extensive benefits but make sustained attention and self-control difficult, especially amid attention-economy incentives. This paper addresses limited understanding by reviewing existing tools through an integrative dual systems framework.

  • INTRODUCTION: Instant access to extensive digital functionality can disrupt current tasks through notifications and habitual check-ins.Many technology business models further encourage frequent, extensive use to optimise advertising revenue.
  • INTRODUCTION: Most users report conflict about time spent with internet-connected technologies and difficulty exercising effective self-control.This has motivated HCI research on intentional non-use and breaks from digital services.
  • INTRODUCTION: Existing understanding remains limited because no systematic review had mapped design features across the hundreds of available digital self-control tools.Prior interventions were mainly informed by user research, design intuition, or theories such as cognitive load, Social Cognitive, and nudge theory.
  • INTRODUCTION: 367 apps and browser extensions are reviewed to identify common design features and intervention strategies for digital self-control.The sample covers the Google Play, Chrome Web, and Apple App stores.
  • INTRODUCTION: The paper adapts an integrative dual systems model incorporating expected value of control to organise and evaluate digital self-control design features.The model is intended to clarify how specific features may scaffold successful self-control.
  • INTRODUCTION: 64% of surveyed smartphone users felt they were overusing their devices, while 60% wanted to change their usage habits.Desired changes involved frequent short use that disrupted focus and excessive long use that absorbed users, while willpower-based strategies often failed.

Theory Applied in Related Research

Prior HCI and digital behaviour-change research has used diverse theories to design or evaluate digital self-control interventions, but dual systems theory was not specified in the reviewed studies. The paper therefore applies an extended dual systems approach to classify existing tools and identify design opportunities.

  • Theory Applied in Related Research: The authors identified 17 HCI papers that built or evaluated interventions supporting self-control over digital device use.They examined which self-regulation theories guided tool development or behavioural and perceptual evaluation.
  • Theory Applied in Related Research: 7 out of 17 papers specified no underlying self-regulation theory, relying only on user-centered design methods.The remaining 10 papers referred to models from psychology, neuroscience, economics, behaviour change, or addiction research.
  • Theory Applied in Related Research: None of the 17 reviewed HCI papers relied on dual systems models of self-regulation.The applied theories included Social Cognitive Theory, classical conditioning, cognitive load, nudge theory, and other models.
  • Theory Applied in Related Research: A review of 85 digital behaviour-change intervention studies found 60% specified no theoretical basis and none specified dual systems theory.The Transtheoretical Model was most common, followed by Goal Setting Theory and Social Conformity Theory.
  • Theory Applied in Related Research: The paper extends dual systems applications in digital behaviour-change interventions with expected value of control to explain fluctuating self-control.It then uses the resulting model to organise common design features in tools from three app and extension stores.

System 1 and System 2

Dual systems theory distinguishes rapid, automatic System 1 processes from slower, conscious, capacity-limited System 2 control. The model explains digital behaviour through competing action schemas shaped by bottom-up inputs and top-down control.

  • System 1 and System 2: System 1 processes are swift, parallel, and non-conscious, whereas System 2 processes are slower, conscious, and capacity-limited.System 2 is needed for planning, decision-making, or overcoming habitual responses and temptations.
  • System 1 and System 2: System 1 maps environmental inputs and internal states to well-learned habits or instinctive responses, including checking smartphone notifications.Such behaviour can begin without conscious awareness and with little interference with other tasks.
  • System 1 and System 2: Action schemas are hierarchical control units for partially ordered action sequences that achieve goals when performed appropriately.They range from simple motor actions to higher-level sequences such as preparing tea.
  • System 1 and System 2: Action schemas compete for behavioural control, with the node having the strongest activation becoming the winner.Activation can come from sensory input, superordinate schemas, or top-down System 2 influence.
  • System 1 and System 2: Figure 1 extends the model by making System 2 control strength depend on the expected value of control.The figure presents System 1 as rapid and non-conscious and System 2 as slower, conscious, and capacity-limited.

Self-regulation and self-control

Self-regulation encompasses goal-directed regulatory processes, while self-control specifically concerns conscious efforts to override conflicting immediate impulses. The framework explains failures and fluctuations through working-memory competition, limited capacity, fatigue, emotional state, and expected value of control.

  • Self-regulation and self-control: Self-regulation includes automatic habits, whereas self-control refers more narrowly to deliberate System 2 efforts against immediate impulses conflicting with valued goals.Both are described as processes serving goal-directed behaviour.
  • Self-regulation and self-control: Goals and intentions must enter working memory before they can guide System 2 control.Attentional filters competitively admit signals with the highest activation values.
  • Self-regulation and self-control: Self-control can fail when relevant goals are not represented in working memory because its capacity is limited.The paper gives the classical estimate of seven, plus or minus two, meaningful information chunks.
  • Self-regulation and self-control: System 2 control fluctuates with fatigue and emotional state, with negative mood predicting relapse and poorer regulation of Facebook use.Continuous exertion can produce fatigue effects, and users regulate platform time less effectively when in a bad mood.
  • Self-regulation and self-control: The expected value of control mediates control strength through perceived reward or avoided loss, expectancy of success, and delay to the outcome.Phone Stack illustrates how a financial and reputational cost can increase the expected value of controlling phone-checking impulses.

A practical example

The example illustrates how habitual checking can override a conscious essay-writing goal when the expected value of control is low, while distracting content can then crowd that goal out of working memory.

  • A student opens his laptop intending to work on an essay but repeatedly checks Facebook and spends too long scrolling its news feed.
  • A habitual laptop context can trigger System 1 checking even when the essay goal is present in working memory.
  • Low expected value of control may prevent System 2 from overriding the checking impulse because inhibition offers little reward or confidence.
  • After Facebook opens, attention-grabbing news-feed content can enter capacity-limited working memory and crowd out the essay goal.

3 A REVIEW AND ANALYSIS OF CURRENT

The paper systematically reviews digital self-control tools and maps their design features onto an adapted dual systems model to examine how they support self-regulation.

  • The review identifies apps and browser extensions that help users exercise self-control, avoid distraction, or manage digital-device-use addiction.
  • The authors adapted an integrative dual systems model to organise and evaluate the coded design features.
  • Searches covered Google Play and Apple App stores using pre-existing scripts and the Chrome Web store using a purpose-built scraper.
  • 4890 distinct apps and extensions remained after duplicate results from multiple search terms and US and UK stores were removed.
  • Reproducibility materials, including data, scripts, and the paper in R Markdown, are available on OSF.

Identifying Potentially Relevant Apps and Extensions.

Screening and coding narrowed the search results to 367 analysable tools, with functionality classified into four anticipated feature clusters.

  • Identifying Potentially Relevant Apps and Extensions.: 731 potentially relevant apps and extensions remained after title and description screening.
  • Identifying Potentially Relevant Apps and Extensions.: Detailed review of descriptions and, when necessary, screenshots reduced the set to 380 apps and extensions.
  • Identifying Potentially Relevant Apps and Extensions.: The Apple version was excluded when an app also existed on Google Play because iOS permissions often limit functionality more than Android permissions.
  • Identifying Potentially Relevant Apps and Extensions.: Functionality was coded from store descriptions, screenshots, and videos using feature categories grouped as block/removal, self-tracking, goal advancement, and reward/punishment.
  • Identifying Potentially Relevant Apps and Extensions.: After iterative independent coding and codebook development, 13 tools were excluded, leaving 367 tools in the final dataset.

Results

Among 367 tools, blocking or removing distractions was most prevalent, while self-tracking, goal advancement, rewards or punishments, and user-defined distraction criteria were also common.

  • 74% of tools included blocking or removing distractions, making it the most frequent feature cluster.
  • 44% (163) created obstacles through blocking, time limits, launch limits, or loading delays, while 14% (50) added friction to removing blocks.
  • 38% (139) reduced exposure to distracting options, mostly through Chrome extensions that removed site elements such as social-media newsfeeds or email inboxes.
  • 38% (139) included self-tracking, commonly recording history, visualising data, or displaying timers and countdowns.
  • 35% (130) supported goal advancement through concrete or general reminders, explicit goals, or comparisons between behaviour and goals.
  • 22% (80) used rewards or punishments, including points, streaks, leaderboards, social sharing, achievements, or virtual lifeforms.
  • 35% (129) let users define distraction, with blacklists more common than whitelists among blocking tools.
  • 65% of tools had one core feature cluster, while 32% (117) combined two, most often block/removal with goal advancement or self-tracking with reward/punishment.

Feature combinations.

Digital self-control tools combine interventions that block or remove distractions, monitor use, advance goals, and influence self-regulation components. Their implementation differs across stores, reflecting developers’ varying control over interfaces and access.

  • HabitLab cycles through intervention types to learn which best helps users align internet use with their stated goals.
  • Browser extensions commonly minimise features, whereas mobile tools more often block or restrict access because developers have less control over other apps’ displays.
  • The model links habit prevention and formation, goal setting and monitoring, rewards, and action-schema competition to distinct design features.
  • Figure 6 reports the percentage of tools containing at least one feature targeting each cognitive component of the dual systems model.
  • Blocking or removing interface features directly prevents unwanted responses by making distracting targets or actions unavailable.

4 DISCUSSION

The review maps the current tool landscape and uses the dual systems model to identify dominant approaches, neglected mechanisms, and opportunities for research and design. It also highlights widely used tools whose mechanisms remain insufficiently evaluated.

  • 4 DISCUSSION: Blocking distractions or removing interface features were the most common approaches, followed by self-tracking, goal advancement, and reward or punishment.
  • 4 DISCUSSION: 65% of tools focused on one design cluster, while 32% focused on two clusters in their core design.
  • 4 DISCUSSION: Feature frequencies differed across stores, likely reflecting differences in developers’ permissions.
  • Research opportunities prompted by widely used or theoretically interesting design features: The market offers hundreds of natural experiments, but many tools lack evaluation of efficacy and transferability of their underlying design mechanisms.
  • Research opportunities prompted by widely used or theoretically interesting design features: Forest links device use to a virtual tree’s well-being, using abstention rather than action to influence the reward component of expected value of control.
  • Research opportunities prompted by widely used or theoretically interesting design features: Timewarp redirects users from distracting websites to productivity-aligned sites, apparently automating implementation intentions that link contexts to desired responses.
  • Research opportunities prompted by widely used or theoretically interesting design features: Tools such as Focusly add commitment by making restrictions difficult to override, raising design and ethical questions about accountability for past preferences.

Gaps identified by the dual systems model

Mapping tools onto the dual systems model identifies underexplored targets: forming desirable habits and influencing delay and expectancy. The model also provides a framework for organising research and guiding theory-aligned intervention design.

  • Gaps identified by the dual systems model: Current tools rarely scaffold new desirable unconscious habits, despite habit formation’s importance for long-term behaviour change.
  • Gaps identified by the dual systems model: Only 4% of tools targeted delay when timer displays were excluded, and just two of 367 tools directly used delays to support self-control.
  • Gaps identified by the dual systems model: Delay is theoretically important because sensitivity to delay has strong, reliable behavioural effects and is central to self-control difficulties.
  • Gaps identified by the dual systems model: Expectancy was infrequently targeted, mainly through timers limiting self-control attempts, despite the role of self-efficacy in self-regulation theory.
  • Gaps identified by the dual systems model: The model can organise interventions by targeted cognitive component and provide a roadmap for studies addressing different parts of the self-regulatory system.
  • Gaps identified by the dual systems model: Its components can expand into more detailed theories about rewards, timing, and gains versus losses to guide design decisions.
  • Gaps identified by the dual systems model: TimeAware illustrates theory-guided design by showing greater productivity support when visualising distracting rather than productive time.

Limitations and future work

The review identifies functionality-focused scope limits and an underspecified cognitive design space, while positioning digital self-control tools as a test bed for future intervention research.

  • Limitations: The analysis covers tool functionality but not install numbers or user-review content, which the authors leave for future work.This scope reflects space restrictions and unavailable Apple App Store user counts.
  • Limitations: The applied dual systems model leaves its cognitive design space underspecified and points to directions for future research.The authors note that more detailed specifications and predictions can be drawn from lower-level theories.
  • Future work: Portable, powerful, internet-connected devices create an unprecedented self-regulation challenge because they make many behavioral options and engaging content instantly available.The same environment also offers opportunities for changing intervention factors at minimal cost.
  • Future work: Digital self-control research can serve as a test bed for interventions that optimize self-control in environments where most factors can be changed at minimal cost.The authors frame this as a broad opportunity for future research.
  • Conclusion: The paper contributes a comprehensive functionality analysis and a self-regulation model intended to provide mechanistic understanding of digital self-control tools.The analysis covers apps and browser extensions from Google Play, Chrome Web, and Apple App stores.
  • Future work: The authors hope the review and dual systems model will support beneficial technology-use habits and resilience against predatory nudging.They describe the reviewed tools as natural experiments in designing for digital self-control.
Loading 1902.00157v1…